Hurricane Risk in a Changing Environment

Eight of the top ten insured property catastrophe losses over the past 38 years have been a result of tropical cyclones striking land (Rudolf et al., 2008). Unfortunately, this year Cyclone Nargis proved once again that landfalling tropical cyclones can also cause massive loss of life and extreme misery. It is not surprising that (re)insurers have an interest knowing as much as possible about how climate affects tropical cyclone activity and the effects of future climate change.

Below I attempt to demonstrate a connection between catastrophe risk models used by (re)insurers to estimate losses and our understanding of climate variability and future climate change. I first summarize the role of catastrophe risk models in capturing our knowledge of how tropical cyclones respond to climate and then discuss the best-track data that are used for model development. I next cover results from analyses of the best-track data showing that on a variety of time scales climate influences tropical cyclone landfall and basin-wide activity. The discussion next moves to a brief review of studies suggesting that tropical cyclone frequency could decrease while at the same time intensity and precipitation could increase. I end with a discussion regarding the relative roles of climate and exposure in the expected future increases in losses from hurricane landfall. The major focus is on hurricanes in the Atlantic that strike the U.S. because these storms produce the largest insured losses and the U.S. is the largest insurance market.

2008. Hurricane risk in a changing environment, in Ponencias de las Jornadas Internacionales Sobre Catastrofes Naturales, v.126 Cuadernos de la Fundacion, Fundacion Mapfre, pp 211-234, 2008. R. Murnane